Effects of Coffea canephora genotypes on the microbial community of soil and fruit Authors: Willian dos Santos Gomes, Fábio Luiz Partelli, Tomás Gomes Reis Veloso, Marliane de Cássia Soares da Silva, Aldemar Polonini Moreli, Taís Rizzo Moreira, Lucas Louzada Pereira Published: 2024 ABSTRACT In recent years, the role of microbial communities in agricultural systems has received increasing attention, particularly concerning their impact on plant health and productivity. However, the influence of host plant genetic factors on the microbial composition of coffee plants remains largely unexplored. This study provides the first comprehensive investigation into how genotype affects the microbial communities present in the rhizosphere and fruits of Coffea canephora. Conducted on a commercial coffee farm in Brazil, we analyzed six genotypes of C. canephora var. Conilon. Soil and fruit samples were collected from which microbial DNA was extracted and sequenced, targeting the V3-V4 region of the 16 S rDNA and the ITS1 region for fungi. A total of 12,239,769 reads were generated from the 16 S rDNA and ITS1 regions. The PCoA revealed distinct patterns of beta diversity, with genotype 153 exhibiting significant isolation in soil bacterial communities. The dominant bacterial orders included Rhizobiales and Rhodobacterales, while the fungal community comprised diverse taxa from Saccharomycetales and Hypocreales. LEfSe analysis identified key metagenomic biomarkers, highlighting genotype Baiano 4 for its richness in fruit-associated taxa, whereas genotype 153 exhibited lower diversity in both soil and fruit samples. This work enhances our understanding of the microbiomes associated with different coffee genotypes, providing evidence of how host genetic variation influences microbial community composition. Our findings indicate that specific microbial taxa are enriched in the fruits and soil of various genotypes. Future research should focus on identifying these microorganisms and elucidating their specific functions within the rhizosphere and coffee fruits.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-024-80403-4. Introduction Plants share an intimate association with microorganisms1. Both bacteria and fungi are found in all plant tissues, where they play distinct roles, either as epiphytes or endophytes2. Numerous studies have demonstrated the direct influence of the microbiota on essential plant functions, including improved protection against pests and diseases, enhanced growth, nutrient acquisition, and increased resistance to biotic and abiotic stresses3–5. However, certain bacteria and fungi associated with plants can act as pathogens, causing diseases and negatively impacting plant performance. In this context, the prevailing understanding within the scientific community is that the microbiota works synergistically with the plant host, strengthening its capacity to effectively adapt to environmental challenges6. The microbiota displays considerable diversity, varying not only between different plant species7but also among genotypes within a species8. Furthermore, this diversity is influenced by the growing environment, indicating that the effects of plant genotypes on their microbiomes are contingent upon the specific cultivation conditions9. Additionally, distinct plant compartments - such as the phyllosphere, rhizosphere, and endosphere – harbor unique microbial communities, each characterized by distinct compositions. Consequently, the assembly of the plant microbiome is largely shaped by the interaction of environmental and genetic factors10. Microbial communities in the rhizosphere soil exhibit considerable variability, both spatially and temporally, resulting in a highly diverse microbial composition that is readily available for colonization11. Entry into the endosphere, which includes the internal spaces of leaves, roots, stems, fruits, and other tissues, requires that microorganisms overcome the plant’s innate immune barriers12. Furthermore, intraspecific genetic variability among different host genotypes profoundly influences the dynamics of microbial communities and ecosystems in multiple ways13. Recent research has significantly expanded our understanding of the interactions between microorganisms and key aspects of coffee’s sensory quality. The diversity of the coffee microbiome varies not only among different regions but also among species, as observed in Brazilian coffee plantations, where specific microbial taxa are associated with Coffea arabica or Coffea canephora, thereby influencing the terroir and overall coffee quality14. The presence and activity of specific groups of microorganisms play crucial roles during coffee processing, such as pectin degradation in the pulp and carbohydrate reduction in the mucilage15. The production of metabolic compounds by these microorganisms can enhance the sensory attributes of coffee beans and the final quality of the beverage16. Although the complexity of this relationship is not yet fully understood, it is important to note that much research on coffee-associated microbiota often concentrates on a single genotype. Plants exhibit substantial genetic variability, particularly in the mechanisms that regulate the plant immune response, as well as in several other phenotypes that have ecological relevance16. Therefore, the differences observed in microbial composition among plant genotypes1 support the hypothesis that genetic variability in coffee plants can significantly affect the selective recruitment of microorganisms that associate with them. Consequently, studies encompassing multiple genotypes are essential to achieve a more comprehensive understanding of the role that host plant genotypes play in the shaping the configuration of the microbiota. Thus, Coffea canephoraemerges as a promising subject for research. Its broad genetic variability is driven by its reproductive nature and self-incompatibility17, as well as the need for genotype diversification to ensure optimal yield and pollination rates18. Moreover, its global relevance is indisputable, contributing approximately 40% to the world’s coffee supply19. It is also important to highlight the limited research on the microbial diversity associated with this species. In light of this, we collected rhizosphere soil and fruit samples from six genotypes of C. canephora, subjecting them to sequencing of the bacterial 16 S and fungal ITS regions to conduct a metataxonomic study. Consequently, we provide a taxonomic and diversity analysis regarding these genotypes. Methods Plant material collection The study was carried out in June 2020 on a commercial coffee farm located in the municipality of Jaguaré, in the northeastern region of the state of Espírito Santo, Brazil (latitude 18°55’42.98”S; longitude 40°10’41.10"W; altitude 93 m). This region is classified under the Köppen climate classification system as Aw, characterized by a tropical wet climate with a dry winter, an average annual temperature of 23.5°C, and an average annual precipitation of 1291 mm. The study examined six genotypes of Coffea canephora var. Conilon, cultivated in rows on a commercial farm. The coffee plants, five years old at the time of the study, were managed following the fertilization guidelines established by Paye et al.20. This management included standard agricultural practices such as weed control, preventive pest and disease management, periodic liming, and drip irrigation, all aimed at ensuring optimal growing conditions and maintaining plant health. The selected genotypes are well-established among local farmers and have been previously studied for their agronomic performance21,22. The LB1 genotype is associated with the “Monte Pascoal” cultivar, which is noted for its high productivity. The 153 genotype belongs to the “Emcapa 8131” cultivar, renowned for its high productivity and large grain size. The A1 genotype, derived from the “Tributun” and “Andina” cultivars, is recognized for its superior sensory quality and adaptation to cultivation at altitudes above 800 m. The VERDIM TA genotype is also linked to the “Andina” cultivar, while the 02 genotype belongs to the “Emcapa 8111” cultivar, which demonstrates promising productive characteristics. Finally, the BAIANO 4 genotype is an emerging genotype that has not yet been associated with a specific cultivar. Collection of fruit and soil samples From each genotype, five composite fruit samples were collected; each sample consisted of 25 fruits from five coffee plants. In addition, five composite soil samples (approximately 500 g each), were obtained from three randomly selected points within a 1 m2 area, at a depth of 10 cm, and beneath the canopy projection of the coffee trees (Fig. 1). Fig. 1Sampling design of soil and fruits in each coffee genotype from three sampling points. The molecular analyses were performed for bacteria and fungi communities. Sampling design of soil and fruits in each coffee genotype from three sampling points. The molecular analyses were performed for bacteria and fungi communities. The samples were placed in plastic bags and stored in the freezer until they were transported to the Laboratory of Mycorrhizal Associations at the Institute of Biotechnology Applied to Agriculture (BIOAGRO), Federal University of Viçosa. Subsequently, they were processed and maintained at −20 °C. DNA extraction and PCR From each composite sample, a total of 250 mg was added to NucleoSpin® Bead Tubes Type A (containing ceramic beads) with 700 µL of buffer SL1 and 150 µL of enhancer SX provided by the Nucleospin Soil® kit (Macherey-Nagel). Cell lysis was performed in a Precellys 24 high-power Bead Mill Homogenizer (Bertin technologies) for 50 s at 4000 rpm. Subsequent steps of DNA extraction were carried out according to the manufacturer’s protocol. The quality of DNA extraction was assessed by electrophoresis on a 0.8% agarose gel stained with ethidium bromide and visualized under UV light. The V3-V4 region of the 16S rRNA gene was amplified using primers 341F (5’-CCTACGGGAGGCAGCAG-3’) and 806R (5’-GGACTACNVGGGTWTCTAAT-3’)23, while the internal Transcribed Spacer 1 (ITS1) of fungi was amplified with primers ITS1F (5’-CTTGGTCATTTAGAGGAAGTAA-3’) and ITS2 (5’-GCTGCGTTCTTCATCGATGC-3’). All PCR reactions were carried out using the Phusion® High-Fidelity PCR Master Mix (New England Biolabs) with specific barcodes for each sample. PCRs products were evaluated on a 2% agarose gel stained with SYBR green. Amplicons ranging from 400 to 450 bp were selected for subsequent experiments. PCR products were pooled in equimolar concentrations and purified using the Qiagen Gel Extraction Kit (Quiagen, Germany). Sequencing libraries were prepared using NEBNext® Ultra™ DNA Library Prep Kit for Illumina, following the manufacturer’s recommendations. Library quality was assessed using a Qubit® 2.0 fluorometer (Thermo Scientific) and the Agilent Bioanalyzer 2100 system. Finally, sequencing was performed on the Illumina NovaSeq 6000 platform, generating paired-end reads of 250 bp pairs. Bioinformatics analysis The raw reads were demultiplexed and truncated to remove primers and adapters. All readings with a maximum number of expected errors (maxee) equal to or greater than one and those identified as chimeras or singletons were removed. The remaining sequences were used to infer Amplicon Sequence Variants (ASVs) using the Amplicon Denoising Algorithm (DADA2)24. The SILVA 138 database was then used to annotate the ASVs. All ASVs classified as organellar DNA (mitochondrial, chloroplast) were excluded from further analyses. All the steps described above were carried out using QIIME2 2021.4 23. Beta diversity of the total microbial community was evaluated based on a Principal coordinate analysis (PCoA) based on the Bray Curtis distance between the samples, using the abundances of 16 S and ITS sequences. This analysis was performed using the microbiome and microeco packages in R Core Team Software25,26packages of R Core Team Software27. Alpha diversity metrics (richness, Pielou equitability, and Shannon diversity) were calculated using the microbiome package25, and differential abundance analyses were conducted with the microeco package28. Values were compared using ANOVA followed by Tuckey post hoc test at a significance level of 0.05. To assess statistical differences in beta and alpha diversity, a Permutational Multivariate Analysis of Variance (PERMANOVA) was conducted, which provides F-statistic values ​​indicating the extent of variation in microbial communities among Coffea canephora genotypes (Table S1). Differential abundance analysis was carried out using the LEfSe (Linear Discriminant Analysis Effect Size) method, as described by Segata et al.29, to identifying high-dimensional biomarkers. LEfSe was applied to detect differentially abundant taxa in the soil and fruits among the Coffea canephora genotypes studied, with a threshold LDA score ≥ 5.0 to highlight the most significant differences. Results A total of 9,229,594 reads were obtained from the sequencing of the V3-V4 region of the 16 S rDNA, while 9,069,731 reads were generated from the ITS1 fungal region. After the removal of chimeras, singletons, and low-quality sequences, 6,877,765 16 S reads and 5,362,004 ITS1 reads were retained for analysis. Beta diversity To assess beta diversity, Principal Coordinate Analysis (PCoA) was performed (Fig. 2). The first two dimensions of PCoA (PC1 and PC2) explained 34% of the total variability in beta diversity of the bacterial and fungal communities in the soil (Fig. 2A and B). In the fruit samples, PC1 and PC2 accounted for 34.7% of the variance in the beta diversity of the bacterial community (Fig. 2C) and 18.1% for the fungal community (Fig. 2D). Fig. 2Principal component analyses displaying the beta diversity of fruit and soil fungal and bacterial communities of six genotypes of C. canephora. Samples of the reference genotype are connected by dashed lines. Principal component analyses displaying the beta diversity of fruit and soil fungal and bacterial communities of six genotypes of C. canephora. Samples of the reference genotype are connected by dashed lines. Principal component analysis revealed a separation in beta diversity of the rhizosphere among different C. canephora genotypes. For the soil bacterial community (Fig. 2A), genotype 153 exhibited a significant differentiation from the other genotypes (Fig. 2A), appearing isolated along the negative axis of PC1. In the soil fungal community, two distinct genotype groups were identified: Group 1, consisting of genotypes 02, A1, and LB1, and Group 2, comprising genotypes Verdim, 153, and Baiano 4. These groups demonstrated markedly different fungal (Fig. 2B). For the bacterial and fungal communities in the fruits of C. canephora genotypes (Fig. 2C and D), although there was distinction in beta diversity between genotypes, the formation of clear clusters, as seen in soil microbial communities, was not observed. Taxonomic composition The study investigated potential differences between the soil and fruit microbiomes of genotypes of Coffea canephora, with the analysis distinguishing between bacteriome and mycobiome (Fig. 3). In the bacterial taxonomic composition of the fruit (Fig. 3A) of Coffea canephora, as observed in the soil, the order Rizhobiales remained predominant across genotypes, except for the A1 genotype, where Rhodobacterales and Sphingobacteriales were observed in one of the samples. Other bacterial taxa identified among the genotypes included Burkholderiales, Micrococcales, Sphingobacteriales, Enterobacterales, and Pseudonocardial. The fungal composition in the fruits (Fig. 3B) showed a diversity of taxa, classified into the orders Saccharomycetales, Hypocreales, Chaetothyriales, Pleosporales, Tremellales, among others. Fig. 3Main orders of bacteria and fungi found in 6 genotypes of C. canephora. Only the most abundant taxa are displayed. The less abundant taxa are shown in the color gray. Main orders of bacteria and fungi found in 6 genotypes of C. canephora. Only the most abundant taxa are displayed. The less abundant taxa are shown in the color gray. Genotype 153 was more closely associated with bacterial taxa that perform essential biogeochemical functions in the soil, such as nitrogen-fixing bacteria belonging to the order Rhizobiales and the family Beijerinckiaceae. Similarly, its fruits were associated with these nitrogen-fixing bacteria (Rhizobiales, Beijerinckiaceae, Methylobacterium) and potential biocontrol agents, represented by the genus Trichoderma. The clone LB1 also exhibited bacteria with nitrogen-fixing potential in both soil and fruit; however, they belonged to a different genus, Sphingononas. Concerning the taxonomic distribution of the soil bacteriome (Fig. 3C), a prevalence of bacteria belonging to the order Rizhobiale was observed in most genotypes. To a lesser extent, the orders Sphingomonadales, Solirubrobacterales, Micromonosporales, Micrococcales, and Burkholderiales were also present. Regarding the soil fungal community (Fig. 3D), a significant proportion of sequences remained unidentified, indicating an opportunity for further explore of this unknown potential. Among the identified taxa, the most predominant were Microascales, Hypocreales, and Pleosporales. Metagenomic biomarkers (LEfSe) To identify the fungal and bacterial taxa in soil and fruit with the most abundant sequences for each C. canephora genotype, a metagenomic biomarker discovery approach (LEfSe) was employed to evaluate the effect sizes of differentially abundant taxa (Fig. 4). Using the LEfSe method, we found that the taxonomic composition (bacteria and fungi) varied significantly between analyzed regions (soil and fruit) and among the C. canephora genotypes. Fig. 4LEfSe analysis of the fungal and bacterial community in soil and fruit of six genotypes of Coffea canephora. The horizontal represents the the Linear Discriminant analysis score of each taxa and the vertical display the taxa names. The color represents the coffee genotype which the microbial taxa is more related. LEfSe analysis of the fungal and bacterial community in soil and fruit of six genotypes of Coffea canephora. The horizontal represents the the Linear Discriminant analysis score of each taxa and the vertical display the taxa names. The color represents the coffee genotype which the microbial taxa is more related. The LDA scores revealed a significant association of biomarkers with the different genotypes analyzed. In the fruit bacterial community, a larger number of taxa were strongly associated with genotype Baiano 4, while no significant markers were detected for genotype 153 (Fig. 4A). Conversely, in the fruit fungal community (Fig. 4B), only the species Septoria cretae was identified as distinctive for genotype Baiano 4 compared to the others. In addition, the LB1 and Baiano 4 genotypes exhibited a significantly higher number of distinctive taxa than the other genotypes. In the soil bacterial community (Fig. 4C), genotypes 153 and Verdim displayed the highest number of distinctive taxa compared to the others. Genotype 02 was strongly associated with the genus Streptomyces, while genotype A1 showed a significant relationship with the order Gemmatimonadales. In the soil fungal community (Fig. 4D), genotypes Baiano 4 and 02 had the highest number of distinctive taxa compared to the other genotypes. For genotype Verdim, taxa belonging to the order Eurotiales were identified as particularly relevant. Alpha diversity Alpha diversity analysis (Fig. 5) was performed to assess species richness and evenness within individual microbial ecosystem, consisting of soil and fruits. Alpha diversity was measured using the Shannon index, a quantitative index that describes the abundance and evenness of species within a host, as opposed to species richness, which solely represents the number of species present. Fig. 5Alpha metrics (richness, evenness, and diversity) of bacteria and fungi present in soil and fruits in genotypes of Coffea canephora. In letters A to D, genotypes with the same letter show no statistical difference at the significance level 0.05. *, * * represent statistical significance at 0.05 and 0.01, respectively, by the Tukey. NS = no significance at the probability level of 0.05. Alpha metrics (richness, evenness, and diversity) of bacteria and fungi present in soil and fruits in genotypes of Coffea canephora. In letters A to D, genotypes with the same letter show no statistical difference at the significance level 0.05. *, * * represent statistical significance at 0.05 and 0.01, respectively, by the Tukey. NS = no significance at the probability level of 0.05. For the bacterial community in fruits, genotype 153 exhibited one of the lowest levels of species richness, as well as lower Shannon entropy (Fig. 5A). However, no significant differences were observed in the evenness of these taxa. Regarding the fungal community in the fruit of the C. canephora genotypes studied (Fig. 5C), no significant differences were detected in the alpha diversity analysis. With respect to the alpha diversity of the microbial community in the soil surrounding the studied genotypes, no significant differences were observed in the Shannon richness and entropy indices for the bacterial community (Fig. 5B). However, genotype 153 displayed lower evenness compared to the other genotypes. As to the fungal community in the soil, no significant differences were found in terms of richness and evenness among genotypes. Nevertheless, genotype 153 had the lowest Shannon entropy index (Fig. 5D). Discussion Beta diversity The observation of distinct beta diversity in the rhizosphere of different of C. canephoragenotypes raises the fundamental question about the mechanisms underlying this phenomenon. One possible explanation is that plants can secrete specific compounds, such as root exudates, which influence the attraction or repulsion of microorganisms25. Studies have shown that these exudates can vary between plant genotypes, playing an important role in each genotype’s ability to attract specific microorganisms and shape their microbial communities26. Therefore, the genotypes distinguished in PCoA may differ in their ability to release these bioactive compounds or in the expression of genes that regulate these interactions. However, the influence of coffee genotypes on the composition of microbial communities in the rhizosphere goes beyond the release of bioactive compounds. The genetic variability of Coffea canephorahas been considered an influential factor in responses to soil conditions, which in turn tends to affect interactions with microorganisms27. Similarly, the specific microbiome composition associated with fruits is highly influenced by intrinsic genetic mechanisms that regulate their interaction with microorganisms, such as the production of antimicrobial compounds and the regulation of the immune response12. These differences based on plant genotypes in the rhizosphere and endophytic microbiota populations have been well demonstrated in several plant species. Studies have reported that even minor variations in the plant genome can significantly alter the structure of microbial communities29. Although variations among genotypes of coffee are still largely unexplored, a plausible hypothesis is that the expression of specific genes for resistance or tolerance to microorganisms varies between the C. canephoragenotypes. Consequently, those genotypes exhibiting genetic resistance to certain pathogens or a selective attraction of beneficial microorganisms could support distinct microbial communities within their fruits13. Taxonomic composition The consistent prevalence of the order Rizhobiales in soil and fruits for most samples raises the question that the presence of this order is little influenced by the C. canephora genotype, and likewise for different C. arabica genotypes30. A possible explanation for this predominance may be the generalist ability of these bacteria to colonize a variety of host plants, regardless of their genetics31. This order includes nitrogen-fixing bacteria, which can be advantageous for plant growth, making them highly adaptable and widely distributed32. In addition, bacteria of this order originated more than 1 billion years ago and have developed a broad genomic plasticity over time, which allows them to adapt to different types of hosts33. Metagenomic biomarkers (LEfSe) In the context of taxonomic composition, the analysis of LDA effect size (LEfSe) (Fig. 4) indicated significant associations between microbial composition and the genotypes examined, emphasizing the distinct characteristics of these genotypes. This analysis suggests a symbiotic relationship between these microbial species and coffee genotypes, where they play crucial roles for the host. The predominance of the fungal genus Trichoderma (Fig. 4B) and the bacterial genus Methylobacterium (Fig. 4A) in fruits associated with 153 and A1 genotypes respectively points to a specific interaction between these genotypes and microorganisms. Trichoderma is recognized for its key role in the control of phytopathogens34, while bacteria of the genus Methylobacterium are associated with nitrogen fixation35. In addition, the presence of another important group of nitrogen fixatives36, as Sphingomonas associated with the genotype LB1 (Fig. 4C), suggests that the microbial composition is influenced by genetic factors unique to this genotype, creating favorable conditions for the growth of these orders. The LEfSe analysis revealed a marked distinction in the taxonomic composition of microorganisms between fruits and soil, which suggests a differential selection of microorganisms due to habitat compartmentalization37. However, this distinction does not exclude the presence of shared taxa between these environments, such as the bacterial orders Rhizobiales and Sphingomonadales, and the fungal orders Hypocreales and Microascales (represented by Scedosporium prolificans), indicating a possible migration and overlap of microbial communities. Furthermore, the LEfSe analysis highlighted that the dynamics of microbial communities can be influenced by the genetic diversity of the host plants, specifically by different coffee genotypes. Distinct host plant genotypes may exhibit genetic variations that shape the composition and dynamics of microbial communities in fruits, leaves, roots, and soil through mechanisms such as root exudation38, production of secondary metabolites39, and differential responses to environmental conditions2. Also, interactions between microorganisms and host plant genotypes can modulate the selection of microorganisms and the mutualistic, pathogenic, or neutral relationships among them40. Alpha diversity When examining the bacterial community in the soil, genotype 153 showed a limited bacterial species richness and a lower evenness compared to other genotypes. This suggests that although there are fewer species, some of them may be present in more significant quantities in this genotype. Despite this lower evenness, this genotype presents the rhizospheric environment with the highest abundance of fixative bacteria of the order Rhizobiales, especially the family Beijerinckiaceae (Fig. 4C). These findings reinforce the premise that the genotypic variation of the host plant favors the growth and predominance of some bacterial species over others. Bulgarelli et al.41 revealed that different genotypes of Arabidopsis exhibit differences in the structure of the bacterial community, reflecting the influence of the genetic characteristics of plants. Thus, the interaction between the plant and soil bacteria can be modulated by genetic factors specific to the host plants. Darriaut et al.42 highlighted the importance of the genetic characteristics of plants in promoting specific microbial communities in the soil, influencing bacterial diversity in the rhizosphere. Therefore, the unique genetic characteristics of C. canephora genotypes can create more favorable microenvironmental conditions for the growth and predominance of certain bacterial species. The absence of significant differences in the richness and evenness of fungal species in fruits among the C. canephoragenotypes suggests that the composition of the fungal community remained relatively constant, regardless of the genotype of the host plant. This phenomenon may be attributed to the role of certain fungal species that create specific microenvironments within the fruits, favoring their survival and predominance. As a result, the variation in fungal species among the different genotypes is minimized43. Conclusion This study investigated the microbial diversity associated with Coffea canephora genotypes, revealing significant differences in bacterial and fungal communities in soils and fruits. Beta diversity analysis showed a clear separation between the genotypes, with genotype 153 standing out for its distinct bacterial composition, rich in nitrogen-fixing bacteria from the Rhizobiales group. In the soil, this order predominated, while fungal diversity varied, with specific groups emerging among genotypes 02, A1, and LB1. The metagenomic biomarkers identified by the LEfSe analysis indicated associations between microbial communities and genotypes, suggesting symbiotic interactions. Alpha diversity revealed that genotype 153 had lower bacterial richness and evenness, although it displayed a significant abundance of nitrogen-fixing bacteria. These results highlight the impact of genetic diversity in Coffea canephora on microbial community formation. Future research should focus on identifying these microorganisms and elucidating their specific functions within the rhizosphere and coffee fruits. Electronic supplementary material Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1